How To Find Foci Of Ellipse

Finding The Foci Of An Ellipse

Most people look at an ellipse and see a stretched circle, which isn't entirely wrong but it misses the whole point of what the shape actually is geometrically. The foci are the anchors. The ellipse is defined by them, not the other way around. That's the first thing most textbooks get backwards and it makes understanding this stuff harder than it needs to be.

How To Find Foci Of Ellipse

The quick version: if your ellipse is centered at the origin and written in standard form, you use c = sqrt(a² - b²), where a is the semi-major axis and b is the semi-minor axis. The foci sit on the major axis, each c units from the center. That's the whole procedure. But how you actually find a and b is where things get messy, depending on what information you're starting with. Let me walk through the three most common scenarios I see people struggle with.

Standard Form From The Start

If you're given an equation like x²/25 + y²/9 = 1, this is already in standard form. The denominator under whichever variable is larger is a². Here, 25 > 9, so a² = 25 and b² = 9. That gives c = sqrt(25 - 9) = sqrt(16) = 4. Since the larger denominator is under x, the major axis is horizontal and the foci are at (±4, 0). But if it were y²/25 + x²/9 = 1, the major axis would be vertical and the foci would be at (0, ±4). The algebra doesn't change. Only the orientation does. People lose points on tests every semester just because they calculated c correctly and then forgot to check which axis was major.

When You're Given The Vertices And Co-vertices

This one comes up more often than you'd think. Say the vertices are at (0, ±7) and the co-vertices are at (±3, 0). The vertices tell you the endpoints of the major axis, so a = 7. The co-vertices are the endpoints of the minor axis, so b = 3. Center is obviously at (0, 0) since the vertices are symmetric around it. Then c = sqrt(49 - 9) = sqrt(40) 6.32. Foci at (0, ±sqrt(40)). I've seen this exact problem show up in AP Calculus classes where students panic because the answer isn't a nice integer. Sqrt(40) is fine. It's still a valid answer. Don't round it unless told to.

The Ugly Case: General Quadratic Form

This is where the real work starts. You're handed something like 25x² + 16y² - 100x + 192y - 156 = 0 and told to find the foci. You have to complete the square on both variables and get it into standard form first. Nobody likes this one. I get that. Group the x terms: 25(x² - 4x). Group the y terms: 16(y² + 12y). Complete the squares inside: you add 4 inside the x group (which is really adding 25 × 4 = 100 to the equation) and 36 inside the y group (which is 16 × 36 = 576). Move those to the right side along with the original constant. You end up with 25(x-2)² + 16(y+6)² = 1024. Divide through by 1024 and you get (x-2)²/40.96 + (y+6)²/64 = 1. Now a² = 64 (under y), b² = 40.96 (under x). Center at (2, -6). Major axis is vertical. c = sqrt(64 - 40.96) = sqrt(23.04) = 4.8. Foci at (2, -6 ± 4.8), which is (2, -1.2) and (2, -10.8).

The arithmetic in this case is tedious but mechanically straightforward. The main failure mode is messing up the completion of the square — specifically forgetting to multiply the added constants by the coefficients outside the parentheses. That mistake alone accounts for roughly half of the incorrect answers I see on this problem type.

What If You're Given The Foci And A Point On The Ellipse

This is the reverse direction and it trips people up differently. The definition of an ellipse is the set of all points where the sum of the distances to the two foci is constant and equal to 2a. So if the foci are at (-3, 0) and (3, 0) and the ellipse passes through (0, 4), you calculate the distance from (0, 4) to each focus, add them together, and that sum is 2a. Distance from (0,4) to (-3,0) is sqrt(9 + 16) = 5. Distance from (0,4) to (3,0) is also sqrt(9 + 16) = 5. Sum is 10, so 2a = 10 and a = 5. The center is midway between the foci at (0,0), and c = 3. Then b² = a² - c² = 25 - 9 = 16. Equation: x²/25 + y²/16 = 1. I ran into a case once where the point given was between the two foci rather than outside them. The distances still add up to 2a, but 2a ended up being less than the distance between the foci themselves. That violates the triangle inequality and means no such ellipse exists. I spent about ten minutes re-reading the problem before catching that the coordinates I'd transcribed had a sign error. Always double-check that a > c before you go further.

A Few Things Nobody Tells You

The eccentricity matters more than you might expect. e = c/a tells you how "stretched" the ellipse is. When e approaches 0, you're basically dealing with a circle and the foci are essentially at the center. When e approaches 1, the foci are near the vertices and the ellipse looks very elongated. In orbital mechanics, for example, planets have e values between 0 and about 0.3 — they're close to circles but the foci shift noticeably over long time scales. If you're working with an ellipse where e is above 0.8, you should question whether calling it an ellipse is even the most useful frame. Rotate conics exist and this method breaks on them. If your general quadratic equation has an xy term (like 7x² + 48xy + 7y² = 200), the major axis isn't aligned with either coordinate axis and the standard-form shortcut doesn't apply. You'd need to rotate the coordinate system by an angle theta where cot(2theta) = (A - C)/B. This comes up in physics problems involving tilted elliptical orbits or stress analysis. It's doable but it's a completely different skill set. I keep a cheat sheet for the rotation angles because I don't trust myself to derive them from memory under time pressure. The sum-of-distances definition is the robust one. Every shortcut formula assumes the ellipse is axis-aligned and centered nicely. When it isn't, falling back on the definition is usually faster than trying to adapt the formula. It also works for any conic section — parabolas and hyperbolas have analogous focal properties. Learning the definition properly saves you from memorizing four different formula variants.

Common Pitfalls

Mixing up a and b is the single most frequent error. Remember: a is always the semi-major axis, regardless of whether it's under x or y. The letter order in the alphabet doesn't correspond to the axis order. Just compare the two denominators and pick the bigger one as a². Another one: forgetting that the foci are always inside the ellipse on the major axis, between the center and the vertices. If your calculated c is larger than a, you made an arithmetic mistake. There is no exception to that rule for real ellipses. A third: treating the co-vertices as if they were foci. They look similar on a diagram — both are points along an axis perpendicular to the major axis. But co-vertices are on the minor axis at distance b from center. Foci are on the major axis at distance c. Mixing these up gives you completely wrong coordinates.

Quick Reference

Standard form with horizontal major axis: (x - h)²/a² + (y - k)²/b² = 1, foci at (h ± c, k) where c = sqrt(a² - b²). Standard form with vertical major axis: (x - h)²/b² + (y - k)²/a² = 1, foci at (h, k ± c) where c = sqrt(a² - b²). Center (h, k) is always the midpoint between the two foci. That's useful for checking your work — if your calculated foci don't have the same midpoint as your calculated vertices, something went wrong somewhere in the process.

I've gone through maybe two dozen different textbook treatments of this topic and they all say roughly the same thing. The ones that waste the most time are the ones that spend pages deriving the standard form from the distance formula before ever mentioning the foci. Skipping straight to c = sqrt(a² - b²) and then explaining where it comes from later is more efficient if you just need to find the foci and move on. The derivation is elegant but it takes about twenty minutes and most people never need to reproduce it.